{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:SCKZ2TKU7R5RXVZCWPLVTA645T","short_pith_number":"pith:SCKZ2TKU","canonical_record":{"source":{"id":"2605.09503","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-10T12:26:50Z","cross_cats_sorted":[],"title_canon_sha256":"cfdfec046d9e408ceb2326dacfe0e3d66d76b14d219b49053b4cf07042f68767","abstract_canon_sha256":"89dd5604c2002a67a55cb5083eb65736dcd9dbddfb035fcb21d6ef526dbfe745"},"schema_version":"1.0"},"canonical_sha256":"90959d4d54fc7b1bd722b3d75983dcecc550c732931eedeef759a807c0dbb2f6","source":{"kind":"arxiv","id":"2605.09503","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.09503","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"arxiv_version","alias_value":"2605.09503v2","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.09503","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_12","alias_value":"SCKZ2TKU7R5R","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_16","alias_value":"SCKZ2TKU7R5RXVZC","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_8","alias_value":"SCKZ2TKU","created_at":"2026-06-02T02:04:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:SCKZ2TKU7R5RXVZCWPLVTA645T","target":"record","payload":{"canonical_record":{"source":{"id":"2605.09503","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-10T12:26:50Z","cross_cats_sorted":[],"title_canon_sha256":"cfdfec046d9e408ceb2326dacfe0e3d66d76b14d219b49053b4cf07042f68767","abstract_canon_sha256":"89dd5604c2002a67a55cb5083eb65736dcd9dbddfb035fcb21d6ef526dbfe745"},"schema_version":"1.0"},"canonical_sha256":"90959d4d54fc7b1bd722b3d75983dcecc550c732931eedeef759a807c0dbb2f6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:18.800007Z","signature_b64":"vz+qit7hdHzI1OfxpwI+QoQ0A3r6TyXUOtC9twMEQvH9tU3KTJV4jFGIpCIcDVSt7p/rHoCNfFdEh9GWtBMJAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"90959d4d54fc7b1bd722b3d75983dcecc550c732931eedeef759a807c0dbb2f6","last_reissued_at":"2026-06-02T02:04:18.799441Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:18.799441Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.09503","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T02:04:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6R7SPSWOCQw4Qnq0QV18SLLGyBr8E+CXVg9ZKuXE5AD/jDXwTHTjnUW8qTmhfKOtCSspecCoc/2iQ+rT1lcdAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T13:35:36.957946Z"},"content_sha256":"d4f36edde35b89fd21d577682eecdc13b652e4266acb8c1cf63da47fca96e716","schema_version":"1.0","event_id":"sha256:d4f36edde35b89fd21d577682eecdc13b652e4266acb8c1cf63da47fca96e716"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:SCKZ2TKU7R5RXVZCWPLVTA645T","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PermuQuant: Lowering Per-Group Quantization Error by Reordering Channels for Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Reordering channels to group similar statistics reduces per-group quantization error in diffusion models.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Junxian Li, Kai Liu, Kaiwen Tao, Renjing Pei, Yongsen Cheng, Yulun Zhang, Zhikai Chen, Zhixin Wang","submitted_at":"2026-05-10T12:26:50Z","abstract_excerpt":"Large-scale visual generative models have achieved remarkable performance. However, their high computational and memory costs make deployment challenging in resource-constrained scenarios, such as interactive applications and personal single-GPU usage. Post-training quantization (PTQ) offers a practical solution by compressing pretrained models without expensive retraining. However, existing PTQ methods still suffer from severe quality degradation under extremely low-bit settings. In this paper, we identify channel ordering as an important but underexplored factor in per-group quantization. In"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"PermuQuant consistently reduces quantization error and outperforms existing PTQ baselines. On FLUX.1-dev with an RTX 5090, PermuQuant achieves up to a 1.8× single step speedup and reduces the DiT memory footprint by 3.5× under W4A4 NVFP4 quantization.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That a permutation chosen on calibration data via the joint second-moment criterion will generalize to the full input distribution at inference time and will not introduce new artifacts in the generated outputs.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"PermuQuant reduces per-group quantization error in diffusion models by sorting channels with similar activation and weight statistics into the same groups using a calibration-checked permutation.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Reordering channels to group similar statistics reduces per-group quantization error in diffusion models.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f4943accd8ed586b19be0e6617004ba1b2428ada61acfacd21b03dbdbf95d9b8"},"source":{"id":"2605.09503","kind":"arxiv","version":2},"verdict":{"id":"8c63ad3d-c07e-4742-b55a-a0c4e3a52aa5","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T04:17:00.312464Z","strongest_claim":"PermuQuant consistently reduces quantization error and outperforms existing PTQ baselines. On FLUX.1-dev with an RTX 5090, PermuQuant achieves up to a 1.8× single step speedup and reduces the DiT memory footprint by 3.5× under W4A4 NVFP4 quantization.","one_line_summary":"PermuQuant reduces per-group quantization error in diffusion models by sorting channels with similar activation and weight statistics into the same groups using a calibration-checked permutation.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That a permutation chosen on calibration data via the joint second-moment criterion will generalize to the full input distribution at inference time and will not introduce new artifacts in the generated outputs.","pith_extraction_headline":"Reordering channels to group similar statistics reduces per-group quantization error in diffusion models."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.09503/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T07:42:01.380920Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T16:41:37.255516Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T13:01:17.787173Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T10:10:27.411584Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"7f6d5d5810d9779f60c7b85a0866222517c7ca8056bd740c810b98e5e21dce8d"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"8c63ad3d-c07e-4742-b55a-a0c4e3a52aa5"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T02:04:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SeyVRl636fmuSpYLcJj4OW2EMICk999CfCrugQQIro+xKR4tjmLGEuVM4yYNDrnJkg7+ahH+j/dGNgP3bPWPDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T13:35:36.958492Z"},"content_sha256":"217678b81dc28d060503f277ca93b0a1fe706b6c82c8fa21d1034ed0f2a873d3","schema_version":"1.0","event_id":"sha256:217678b81dc28d060503f277ca93b0a1fe706b6c82c8fa21d1034ed0f2a873d3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SCKZ2TKU7R5RXVZCWPLVTA645T/bundle.json","state_url":"https://pith.science/pith/SCKZ2TKU7R5RXVZCWPLVTA645T/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SCKZ2TKU7R5RXVZCWPLVTA645T/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-04T13:35:36Z","links":{"resolver":"https://pith.science/pith/SCKZ2TKU7R5RXVZCWPLVTA645T","bundle":"https://pith.science/pith/SCKZ2TKU7R5RXVZCWPLVTA645T/bundle.json","state":"https://pith.science/pith/SCKZ2TKU7R5RXVZCWPLVTA645T/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SCKZ2TKU7R5RXVZCWPLVTA645T/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SCKZ2TKU7R5RXVZCWPLVTA645T","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"89dd5604c2002a67a55cb5083eb65736dcd9dbddfb035fcb21d6ef526dbfe745","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-10T12:26:50Z","title_canon_sha256":"cfdfec046d9e408ceb2326dacfe0e3d66d76b14d219b49053b4cf07042f68767"},"schema_version":"1.0","source":{"id":"2605.09503","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.09503","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"arxiv_version","alias_value":"2605.09503v2","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.09503","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_12","alias_value":"SCKZ2TKU7R5R","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_16","alias_value":"SCKZ2TKU7R5RXVZC","created_at":"2026-06-02T02:04:18Z"},{"alias_kind":"pith_short_8","alias_value":"SCKZ2TKU","created_at":"2026-06-02T02:04:18Z"}],"graph_snapshots":[{"event_id":"sha256:217678b81dc28d060503f277ca93b0a1fe706b6c82c8fa21d1034ed0f2a873d3","target":"graph","created_at":"2026-06-02T02:04:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"PermuQuant consistently reduces quantization error and outperforms existing PTQ baselines. On FLUX.1-dev with an RTX 5090, PermuQuant achieves up to a 1.8× single step speedup and reduces the DiT memory footprint by 3.5× under W4A4 NVFP4 quantization."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That a permutation chosen on calibration data via the joint second-moment criterion will generalize to the full input distribution at inference time and will not introduce new artifacts in the generated outputs."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"PermuQuant reduces per-group quantization error in diffusion models by sorting channels with similar activation and weight statistics into the same groups using a calibration-checked permutation."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Reordering channels to group similar statistics reduces per-group quantization error in diffusion models."}],"snapshot_sha256":"f4943accd8ed586b19be0e6617004ba1b2428ada61acfacd21b03dbdbf95d9b8"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T07:42:01.380920Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T16:41:37.255516Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T13:01:17.787173Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T10:10:27.411584Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.09503/integrity.json","findings":[],"snapshot_sha256":"7f6d5d5810d9779f60c7b85a0866222517c7ca8056bd740c810b98e5e21dce8d","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large-scale visual generative models have achieved remarkable performance. However, their high computational and memory costs make deployment challenging in resource-constrained scenarios, such as interactive applications and personal single-GPU usage. Post-training quantization (PTQ) offers a practical solution by compressing pretrained models without expensive retraining. However, existing PTQ methods still suffer from severe quality degradation under extremely low-bit settings. In this paper, we identify channel ordering as an important but underexplored factor in per-group quantization. In","authors_text":"Junxian Li, Kai Liu, Kaiwen Tao, Renjing Pei, Yongsen Cheng, Yulun Zhang, Zhikai Chen, Zhixin Wang","cross_cats":[],"headline":"Reordering channels to group similar statistics reduces per-group quantization error in diffusion models.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-10T12:26:50Z","title":"PermuQuant: Lowering Per-Group Quantization Error by Reordering Channels for Diffusion Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.09503","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-12T04:17:00.312464Z","id":"8c63ad3d-c07e-4742-b55a-a0c4e3a52aa5","model_set":{"reader":"grok-4.3"},"one_line_summary":"PermuQuant reduces per-group quantization error in diffusion models by sorting channels with similar activation and weight statistics into the same groups using a calibration-checked permutation.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Reordering channels to group similar statistics reduces per-group quantization error in diffusion models.","strongest_claim":"PermuQuant consistently reduces quantization error and outperforms existing PTQ baselines. On FLUX.1-dev with an RTX 5090, PermuQuant achieves up to a 1.8× single step speedup and reduces the DiT memory footprint by 3.5× under W4A4 NVFP4 quantization.","weakest_assumption":"That a permutation chosen on calibration data via the joint second-moment criterion will generalize to the full input distribution at inference time and will not introduce new artifacts in the generated outputs."}},"verdict_id":"8c63ad3d-c07e-4742-b55a-a0c4e3a52aa5"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d4f36edde35b89fd21d577682eecdc13b652e4266acb8c1cf63da47fca96e716","target":"record","created_at":"2026-06-02T02:04:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"89dd5604c2002a67a55cb5083eb65736dcd9dbddfb035fcb21d6ef526dbfe745","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-10T12:26:50Z","title_canon_sha256":"cfdfec046d9e408ceb2326dacfe0e3d66d76b14d219b49053b4cf07042f68767"},"schema_version":"1.0","source":{"id":"2605.09503","kind":"arxiv","version":2}},"canonical_sha256":"90959d4d54fc7b1bd722b3d75983dcecc550c732931eedeef759a807c0dbb2f6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"90959d4d54fc7b1bd722b3d75983dcecc550c732931eedeef759a807c0dbb2f6","first_computed_at":"2026-06-02T02:04:18.799441Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T02:04:18.799441Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vz+qit7hdHzI1OfxpwI+QoQ0A3r6TyXUOtC9twMEQvH9tU3KTJV4jFGIpCIcDVSt7p/rHoCNfFdEh9GWtBMJAw==","signature_status":"signed_v1","signed_at":"2026-06-02T02:04:18.800007Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.09503","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d4f36edde35b89fd21d577682eecdc13b652e4266acb8c1cf63da47fca96e716","sha256:217678b81dc28d060503f277ca93b0a1fe706b6c82c8fa21d1034ed0f2a873d3"],"state_sha256":"77498c3d856e58e996f259b4592a128d0a709a706efda2f10f3956f67b944890"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IakFaMu/M4JYPmQdL4CzAPVKcEQjFLv+7xwh/EB4PeVx95Tv/r5IhvLlBqYeUOnYWkyWH/05bDLnIPbaqNg2Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T13:35:36.960876Z","bundle_sha256":"6cef28c4f493798eb294361ccca9debb576e20fb7788c0684276225e98af1c5b"}}